gmakstutis/graphiti
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Remove ensure_ascii configuration parameter

Browse source 
- Changed to_prompt_json default from ensure_ascii=True to False
- Removed ensure_ascii parameter from Graphiti.__init__ and GraphitiClients
- Removed ensure_ascii from all function signatures and context dictionaries
- Removed ensure_ascii from all test files
- All JSON serialization now preserves Unicode characters by default

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Daniel Chalef 2025-10-02 15:03:03 -07:00
parent bec3f02036
commit 5ab8eee576
18 changed files with 47 additions and 103 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
graphiti_core/graphiti.py
View file

@ -136,7 +136,6 @@ class Graphiti:
store_raw_episode_content: bool = True,
graph_driver: GraphDriver | None = None,
max_coroutines: int | None = None,
ensure_ascii: bool = False,
):
"""
Initialize a Graphiti instance.
@ -169,10 +168,6 @@ class Graphiti:
max_coroutines : int | None, optional
The maximum number of concurrent operations allowed. Overrides SEMAPHORE_LIMIT set in the environment.
If not set, the Graphiti default is used.
ensure_ascii : bool, optional
Whether to escape non-ASCII characters in JSON serialization for prompts. Defaults to False.
Set as False to preserve non-ASCII characters (e.g., Korean, Japanese, Chinese) in their
original form, making them readable in LLM logs and improving model understanding.
Returns
-------
@ -202,7 +197,6 @@ class Graphiti:
self.store_raw_episode_content = store_raw_episode_content
self.max_coroutines = max_coroutines
self.ensure_ascii = ensure_ascii
if llm_client:
self.llm_client = llm_client
else:
@ -221,7 +215,6 @@ class Graphiti:
llm_client=self.llm_client,
embedder=self.embedder,
cross_encoder=self.cross_encoder,
ensure_ascii=self.ensure_ascii,
)
# Capture telemetry event
@ -560,7 +553,7 @@ class Graphiti:
communities, community_edges = await semaphore_gather(
*[
update_community(
self.driver, self.llm_client, self.embedder, node, self.ensure_ascii
self.driver, self.llm_client, self.embedder, node
)
for node in nodes
],
@ -1071,7 +1064,6 @@ class Graphiti:
),
None,
None,
self.ensure_ascii,
)
edges: list[EntityEdge] = [resolved_edge] + invalidated_edges

1
graphiti_core/graphiti_types.py
View file

@ -27,6 +27,5 @@ class GraphitiClients(BaseModel):
llm_client: LLMClient
embedder: EmbedderClient
cross_encoder: CrossEncoderClient
ensure_ascii: bool = False
model_config = ConfigDict(arbitrary_types_allowed=True)

6
graphiti_core/prompts/dedupe_edges.py
View file

@ -67,11 +67,11 @@ def edge(context: dict[str, Any]) -> list[Message]:
Given the following context, determine whether the New Edge represents any of the edges in the list of Existing Edges.
<EXISTING EDGES>
{to_prompt_json(context['related_edges'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['related_edges'], indent=2)}
</EXISTING EDGES>
<NEW EDGE>
{to_prompt_json(context['extracted_edges'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['extracted_edges'], indent=2)}
</NEW EDGE>
Task:
@ -98,7 +98,7 @@ def edge_list(context: dict[str, Any]) -> list[Message]:
Given the following context, find all of the duplicates in a list of facts:
Facts:
{to_prompt_json(context['edges'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['edges'], indent=2)}
Task:
If any facts in Facts is a duplicate of another fact, return a new fact with one of their uuid's.

16
graphiti_core/prompts/dedupe_nodes.py
View file

@ -64,20 +64,20 @@ def node(context: dict[str, Any]) -> list[Message]:
role='user',
content=f"""
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
{context['episode_content']}
</CURRENT MESSAGE>
<NEW ENTITY>
{to_prompt_json(context['extracted_node'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['extracted_node'], indent=2)}
</NEW ENTITY>
<ENTITY TYPE DESCRIPTION>
{to_prompt_json(context['entity_type_description'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['entity_type_description'], indent=2)}
</ENTITY TYPE DESCRIPTION>
<EXISTING ENTITIES>
{to_prompt_json(context['existing_nodes'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['existing_nodes'], indent=2)}
</EXISTING ENTITIES>
Given the above EXISTING ENTITIES and their attributes, MESSAGE, and PREVIOUS MESSAGES; Determine if the NEW ENTITY extracted from the conversation
@ -125,7 +125,7 @@ def nodes(context: dict[str, Any]) -> list[Message]:
role='user',
content=f"""
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
{context['episode_content']}
@ -142,11 +142,11 @@ def nodes(context: dict[str, Any]) -> list[Message]:
}}
<ENTITIES>
{to_prompt_json(context['extracted_nodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['extracted_nodes'], indent=2)}
</ENTITIES>
<EXISTING ENTITIES>
{to_prompt_json(context['existing_nodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['existing_nodes'], indent=2)}
</EXISTING ENTITIES>
Each entry in EXISTING ENTITIES is an object with the following structure:
@ -197,7 +197,7 @@ def node_list(context: dict[str, Any]) -> list[Message]:
Given the following context, deduplicate a list of nodes:
Nodes:
{to_prompt_json(context['nodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['nodes'], indent=2)}
Task:
1. Group nodes together such that all duplicate nodes are in the same list of uuids

6
graphiti_core/prompts/eval.py
View file

@ -68,7 +68,7 @@ def query_expansion(context: dict[str, Any]) -> list[Message]:
Bob is asking Alice a question, are you able to rephrase the question into a simpler one about Alice in the third person
that maintains the relevant context?
<QUESTION>
{to_prompt_json(context['query'], ensure_ascii=context.get('ensure_ascii', False))}
{to_prompt_json(context['query'])}
</QUESTION>
"""
return [
@ -84,10 +84,10 @@ def qa_prompt(context: dict[str, Any]) -> list[Message]:
Your task is to briefly answer the question in the way that you think Alice would answer the question.
You are given the following entity summaries and facts to help you determine the answer to your question.
<ENTITY_SUMMARIES>
{to_prompt_json(context['entity_summaries'], ensure_ascii=context.get('ensure_ascii', False))}
{to_prompt_json(context['entity_summaries'])}
</ENTITY_SUMMARIES>
<FACTS>
{to_prompt_json(context['facts'], ensure_ascii=context.get('ensure_ascii', False))}
{to_prompt_json(context['facts'])}
</FACTS>
<QUESTION>
{context['query']}

6
graphiti_core/prompts/extract_edges.py
View file

@ -73,7 +73,7 @@ def edge(context: dict[str, Any]) -> list[Message]:
</FACT TYPES>
<PREVIOUS_MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS_MESSAGES>
<CURRENT_MESSAGE>
@ -133,7 +133,7 @@ def reflexion(context: dict[str, Any]) -> list[Message]:
user_prompt = f"""
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
{context['episode_content']}
@ -167,7 +167,7 @@ def extract_attributes(context: dict[str, Any]) -> list[Message]:
content=f"""
<MESSAGE>
{to_prompt_json(context['episode_content'], ensure_ascii=context.get('ensure_ascii', False), indent=2)}
{to_prompt_json(context['episode_content'], indent=2)}
</MESSAGE>
<REFERENCE TIME>
{context['reference_time']}

14
graphiti_core/prompts/extract_nodes.py
View file

@ -89,7 +89,7 @@ def extract_message(context: dict[str, Any]) -> list[Message]:
</ENTITY TYPES>
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
@ -197,7 +197,7 @@ def reflexion(context: dict[str, Any]) -> list[Message]:
user_prompt = f"""
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
{context['episode_content']}
@ -221,7 +221,7 @@ def classify_nodes(context: dict[str, Any]) -> list[Message]:
user_prompt = f"""
<PREVIOUS MESSAGES>
{to_prompt_json([ep for ep in context['previous_episodes']], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json([ep for ep in context['previous_episodes']], indent=2)}
</PREVIOUS MESSAGES>
<CURRENT MESSAGE>
{context['episode_content']}
@ -259,8 +259,8 @@ def extract_attributes(context: dict[str, Any]) -> list[Message]:
content=f"""
<MESSAGES>
{to_prompt_json(context['previous_episodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['episode_content'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['previous_episodes'], indent=2)}
{to_prompt_json(context['episode_content'], indent=2)}
</MESSAGES>
Given the above MESSAGES and the following ENTITY, update any of its attributes based on the information provided
@ -289,8 +289,8 @@ def extract_summary(context: dict[str, Any]) -> list[Message]:
content=f"""
<MESSAGES>
{to_prompt_json(context['previous_episodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['episode_content'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['previous_episodes'], indent=2)}
{to_prompt_json(context['episode_content'], indent=2)}
</MESSAGES>
Given the above MESSAGES and the following ENTITY, update the summary that combines relevant information about the entity

6
graphiti_core/prompts/prompt_helpers.py
View file

@ -4,20 +4,20 @@ from typing import Any
DO_NOT_ESCAPE_UNICODE = '\nDo not escape unicode characters.\n'
def to_prompt_json(data: Any, ensure_ascii: bool = True, indent: int = 2) -> str:
def to_prompt_json(data: Any, ensure_ascii: bool = False, indent: int = 2) -> str:
"""
Serialize data to JSON for use in prompts.
Args:
data: The data to serialize
ensure_ascii: If True, escape non-ASCII characters. If False, preserve them.
ensure_ascii: If True, escape non-ASCII characters. If False (default), preserve them.
indent: Number of spaces for indentation
Returns:
JSON string representation of the data
Notes:
When ensure_ascii=False, non-ASCII characters (e.g., Korean, Japanese, Chinese)
By default (ensure_ascii=False), non-ASCII characters (e.g., Korean, Japanese, Chinese)
are preserved in their original form in the prompt, making them readable
in LLM logs and improving model understanding.
"""

10
graphiti_core/prompts/summarize_nodes.py
View file

@ -59,7 +59,7 @@ def summarize_pair(context: dict[str, Any]) -> list[Message]:
Summaries must be under 250 words.
Summaries:
{to_prompt_json(context['node_summaries'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['node_summaries'], indent=2)}
""",
),
]
@ -76,8 +76,8 @@ def summarize_context(context: dict[str, Any]) -> list[Message]:
content=f"""
<MESSAGES>
{to_prompt_json(context['previous_episodes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['episode_content'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['previous_episodes'], indent=2)}
{to_prompt_json(context['episode_content'], indent=2)}
</MESSAGES>
Given the above MESSAGES and the following ENTITY name, create a summary for the ENTITY. Your summary must only use
@ -100,7 +100,7 @@ def summarize_context(context: dict[str, Any]) -> list[Message]:
</ENTITY CONTEXT>
<ATTRIBUTES>
{to_prompt_json(context['attributes'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['attributes'], indent=2)}
</ATTRIBUTES>
""",
),
@ -120,7 +120,7 @@ def summary_description(context: dict[str, Any]) -> list[Message]:
Summaries must be under 250 words.
Summary:
{to_prompt_json(context['summary'], ensure_ascii=context.get('ensure_ascii', True), indent=2)}
{to_prompt_json(context['summary'], indent=2)}
""",
),
]

10
graphiti_core/search/search_helpers.py
View file

@ -25,7 +25,7 @@ def format_edge_date_range(edge: EntityEdge) -> str:
def search_results_to_context_string(
search_results: SearchResults, ensure_ascii: bool = False
search_results: SearchResults
) -> str:
"""Reformats a set of SearchResults into a single string to pass directly to an LLM as context"""
fact_json = [
@ -58,16 +58,16 @@ def search_results_to_context_string(
These are the most relevant facts and their valid and invalid dates. Facts are considered valid
between their valid_at and invalid_at dates. Facts with an invalid_at date of "Present" are considered valid.
<FACTS>
{to_prompt_json(fact_json, ensure_ascii=ensure_ascii, indent=12)}
{to_prompt_json(fact_json, indent=12)}
</FACTS>
<ENTITIES>
{to_prompt_json(entity_json, ensure_ascii=ensure_ascii, indent=12)}
{to_prompt_json(entity_json, indent=12)}
</ENTITIES>
<EPISODES>
{to_prompt_json(episode_json, ensure_ascii=ensure_ascii, indent=12)}
{to_prompt_json(episode_json, indent=12)}
</EPISODES>
<COMMUNITIES>
{to_prompt_json(community_json, ensure_ascii=ensure_ascii, indent=12)}
{to_prompt_json(community_json, indent=12)}
</COMMUNITIES>
"""

1
graphiti_core/utils/bulk_utils.py
View file

@ -479,7 +479,6 @@ async def dedupe_edges_bulk(
episode,
edge_types,
set(edge_types),
clients.ensure_ascii,
)
for episode, edge, candidates in dedupe_tuples
]

20
graphiti_core/utils/maintenance/community_operations.py
View file

@ -132,12 +132,11 @@ def label_propagation(projection: dict[str, list[Neighbor]]) -> list[list[str]]:
async def summarize_pair(
llm_client: LLMClient, summary_pair: tuple[str, str], ensure_ascii: bool = True
llm_client: LLMClient, summary_pair: tuple[str, str]
) -> str:
# Prepare context for LLM
context = {
'node_summaries': [{'summary': summary} for summary in summary_pair],
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
@ -150,11 +149,10 @@ async def summarize_pair(
async def generate_summary_description(
llm_client: LLMClient, summary: str, ensure_ascii: bool = True
llm_client: LLMClient, summary: str
) -> str:
context = {
'summary': summary,
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
@ -168,7 +166,7 @@ async def generate_summary_description(
async def build_community(
llm_client: LLMClient, community_cluster: list[EntityNode], ensure_ascii: bool = True
llm_client: LLMClient, community_cluster: list[EntityNode]
) -> tuple[CommunityNode, list[CommunityEdge]]:
summaries = [entity.summary for entity in community_cluster]
length = len(summaries)
@ -181,7 +179,7 @@ async def build_community(
await semaphore_gather(
*[
summarize_pair(
llm_client, (str(left_summary), str(right_summary)), ensure_ascii
llm_client, (str(left_summary), str(right_summary))
)
for left_summary, right_summary in zip(
summaries[: int(length / 2)], summaries[int(length / 2) :], strict=False
@ -195,7 +193,7 @@ async def build_community(
length = len(summaries)
summary = summaries[0]
name = await generate_summary_description(llm_client, summary, ensure_ascii)
name = await generate_summary_description(llm_client, summary)
now = utc_now()
community_node = CommunityNode(
name=name,
@ -215,7 +213,6 @@ async def build_communities(
driver: GraphDriver,
llm_client: LLMClient,
group_ids: list[str] | None,
ensure_ascii: bool = True,
) -> tuple[list[CommunityNode], list[CommunityEdge]]:
community_clusters = await get_community_clusters(driver, group_ids)
@ -223,7 +220,7 @@ async def build_communities(
async def limited_build_community(cluster):
async with semaphore:
return await build_community(llm_client, cluster, ensure_ascii)
return await build_community(llm_client, cluster)
communities: list[tuple[CommunityNode, list[CommunityEdge]]] = list(
await semaphore_gather(
@ -312,7 +309,6 @@ async def update_community(
llm_client: LLMClient,
embedder: EmbedderClient,
entity: EntityNode,
ensure_ascii: bool = True,
) -> tuple[list[CommunityNode], list[CommunityEdge]]:
community, is_new = await determine_entity_community(driver, entity)
@ -320,9 +316,9 @@ async def update_community(
return [], []
new_summary = await summarize_pair(
llm_client, (entity.summary, community.summary), ensure_ascii
llm_client, (entity.summary, community.summary)
)
new_name = await generate_summary_description(llm_client, new_summary, ensure_ascii)
new_name = await generate_summary_description(llm_client, new_summary)
community.summary = new_summary
community.name = new_name

9
graphiti_core/utils/maintenance/edge_operations.py
View file

@ -130,7 +130,6 @@ async def extract_edges(
'reference_time': episode.valid_at,
'edge_types': edge_types_context,
'custom_prompt': '',
'ensure_ascii': clients.ensure_ascii,
}
facts_missed = True
@ -358,7 +357,6 @@ async def resolve_extracted_edges(
episode,
extracted_edge_types,
custom_type_names,
clients.ensure_ascii,
)
for extracted_edge, related_edges, existing_edges, extracted_edge_types in zip(
extracted_edges,
@ -431,7 +429,6 @@ async def resolve_extracted_edge(
episode: EpisodicNode,
edge_type_candidates: dict[str, type[BaseModel]] | None = None,
custom_edge_type_names: set[str] | None = None,
ensure_ascii: bool = True,
) -> tuple[EntityEdge, list[EntityEdge], list[EntityEdge]]:
"""Resolve an extracted edge against existing graph context.
@ -453,8 +450,6 @@ async def resolve_extracted_edge(
Full catalog of registered custom edge names. Used to distinguish
between disallowed custom types (which fall back to the default label)
and ad-hoc labels emitted by the LLM.
ensure_ascii : bool
Whether prompt payloads should coerce ASCII output.
Returns
-------
@ -504,7 +499,6 @@ async def resolve_extracted_edge(
'new_edge': extracted_edge.fact,
'edge_invalidation_candidates': invalidation_edge_candidates_context,
'edge_types': edge_types_context,
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
@ -548,8 +542,7 @@ async def resolve_extracted_edge(
'episode_content': episode.content,
'reference_time': episode.valid_at,
'fact': resolved_edge.fact,
'ensure_ascii': ensure_ascii,
}
}
edge_model = edge_type_candidates.get(fact_type) if edge_type_candidates else None
if edge_model is not None and len(edge_model.model_fields) != 0:

11
graphiti_core/utils/maintenance/node_operations.py
View file

@ -64,14 +64,12 @@ async def extract_nodes_reflexion(
episode: EpisodicNode,
previous_episodes: list[EpisodicNode],
node_names: list[str],
ensure_ascii: bool = False,
) -> list[str]:
# Prepare context for LLM
context = {
'episode_content': episode.content,
'previous_episodes': [ep.content for ep in previous_episodes],
'extracted_entities': node_names,
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
@ -124,7 +122,6 @@ async def extract_nodes(
'custom_prompt': custom_prompt,
'entity_types': entity_types_context,
'source_description': episode.source_description,
'ensure_ascii': clients.ensure_ascii,
}
while entities_missed and reflexion_iterations <= MAX_REFLEXION_ITERATIONS:
@ -155,7 +152,6 @@ async def extract_nodes(
episode,
previous_episodes,
[entity.name for entity in extracted_entities],
clients.ensure_ascii,
)
entities_missed = len(missing_entities) != 0
@ -239,7 +235,6 @@ async def _resolve_with_llm(
extracted_nodes: list[EntityNode],
indexes: DedupCandidateIndexes,
state: DedupResolutionState,
ensure_ascii: bool,
episode: EpisodicNode | None,
previous_episodes: list[EpisodicNode] | None,
entity_types: dict[str, type[BaseModel]] | None,
@ -309,7 +304,6 @@ async def _resolve_with_llm(
'previous_episodes': (
[ep.content for ep in previous_episodes] if previous_episodes is not None else []
),
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
@ -416,7 +410,6 @@ async def resolve_extracted_nodes(
extracted_nodes,
indexes,
state,
clients.ensure_ascii,
episode,
previous_episodes,
entity_types,
@ -465,7 +458,6 @@ async def extract_attributes_from_nodes(
if entity_types is not None
else None
),
clients.ensure_ascii,
should_summarize_node,
)
for node in nodes
@ -483,7 +475,6 @@ async def extract_attributes_from_node(
episode: EpisodicNode | None = None,
previous_episodes: list[EpisodicNode] | None = None,
entity_type: type[BaseModel] | None = None,
ensure_ascii: bool = False,
should_summarize_node: NodeSummaryFilter | None = None,
) -> EntityNode:
node_context: dict[str, Any] = {
@ -499,7 +490,6 @@ async def extract_attributes_from_node(
'previous_episodes': (
[ep.content for ep in previous_episodes] if previous_episodes is not None else []
),
'ensure_ascii': ensure_ascii,
}
summary_context: dict[str, Any] = {
@ -508,7 +498,6 @@ async def extract_attributes_from_node(
'previous_episodes': (
[ep.content for ep in previous_episodes] if previous_episodes is not None else []
),
'ensure_ascii': ensure_ascii,
}
has_entity_attributes: bool = bool(

4
graphiti_core/utils/maintenance/temporal_operations.py
View file

@ -35,14 +35,12 @@ async def extract_edge_dates(
edge: EntityEdge,
current_episode: EpisodicNode,
previous_episodes: list[EpisodicNode],
ensure_ascii: bool = False,
) -> tuple[datetime | None, datetime | None]:
context = {
'edge_fact': edge.fact,
'current_episode': current_episode.content,
'previous_episodes': [ep.content for ep in previous_episodes],
'reference_timestamp': current_episode.valid_at.isoformat(),
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(
prompt_library.extract_edge_dates.v1(context), response_model=EdgeDates
@ -75,7 +73,6 @@ async def get_edge_contradictions(
llm_client: LLMClient,
new_edge: EntityEdge,
existing_edges: list[EntityEdge],
ensure_ascii: bool = False,
) -> list[EntityEdge]:
start = time()
@ -87,7 +84,6 @@ async def get_edge_contradictions(
context = {
'new_edge': new_edge_context,
'existing_edges': existing_edge_context,
'ensure_ascii': ensure_ascii,
}
llm_response = await llm_client.generate_response(

2
tests/utils/maintenance/test_bulk_utils.py
View file

@ -34,7 +34,6 @@ def _make_clients() -> GraphitiClients:
embedder=embedder,
cross_encoder=cross_encoder,
llm_client=llm_client,
ensure_ascii=False,
)
@ -260,7 +259,6 @@ async def test_dedupe_edges_bulk_deduplicates_within_episode(monkeypatch):
episode,
edge_type_candidates=None,
custom_edge_type_names=None,
ensure_ascii=False,
):
# Track that this edge was compared against the related_edges
comparisons_made.append((extracted_edge.uuid, [r.uuid for r in related_edges]))

8
tests/utils/maintenance/test_edge_operations.py
View file

@ -143,7 +143,6 @@ async def test_resolve_extracted_edge_exact_fact_short_circuit(
mock_existing_edges,
mock_current_episode,
edge_type_candidates=None,
ensure_ascii=True,
)
assert resolved_edge is related_edges[0]
@ -184,7 +183,6 @@ async def test_resolve_extracted_edges_resets_unmapped_names(monkeypatch):
llm_client=llm_client,
embedder=MagicMock(),
cross_encoder=MagicMock(),
ensure_ascii=True,
)
source_node = EntityNode(
@ -265,7 +263,6 @@ async def test_resolve_extracted_edges_keeps_unknown_names(monkeypatch):
llm_client=llm_client,
embedder=MagicMock(),
cross_encoder=MagicMock(),
ensure_ascii=True,
)
source_node = EntityNode(
@ -369,7 +366,6 @@ async def test_resolve_extracted_edge_rejects_unmapped_fact_type(mock_llm_client
episode,
edge_type_candidates={},
custom_edge_type_names={'OCCURRED_AT'},
ensure_ascii=True,
)
assert resolved_edge.name == DEFAULT_EDGE_NAME
@ -427,7 +423,6 @@ async def test_resolve_extracted_edge_accepts_unknown_fact_type(mock_llm_client)
episode,
edge_type_candidates={'OCCURRED_AT': OccurredAtEdge},
custom_edge_type_names={'OCCURRED_AT'},
ensure_ascii=True,
)
assert resolved_edge.name == 'INTERACTED_WITH'
@ -515,7 +510,6 @@ async def test_resolve_extracted_edge_uses_integer_indices_for_duplicates(mock_l
episode,
edge_type_candidates=None,
custom_edge_type_names=set(),
ensure_ascii=True,
)
# Verify LLM was called
@ -553,7 +547,6 @@ async def test_resolve_extracted_edges_fast_path_deduplication(monkeypatch):
episode,
edge_type_candidates=None,
custom_edge_type_names=None,
ensure_ascii=False,
):
nonlocal resolve_call_count
resolve_call_count += 1
@ -576,7 +569,6 @@ async def test_resolve_extracted_edges_fast_path_deduplication(monkeypatch):
llm_client=llm_client,
embedder=MagicMock(),
cross_encoder=MagicMock(),
ensure_ascii=True,
)
source_node = EntityNode(

10
tests/utils/maintenance/test_node_operations.py
View file

@ -46,7 +46,6 @@ def _make_clients():
embedder=embedder,
cross_encoder=cross_encoder,
llm_client=llm_client,
ensure_ascii=False,
)
return clients, llm_generate
@ -335,7 +334,6 @@ async def test_resolve_with_llm_updates_unresolved(monkeypatch):
[extracted],
indexes,
state,
ensure_ascii=False,
episode=_make_episode(),
previous_episodes=[],
entity_types=None,
@ -380,7 +378,6 @@ async def test_resolve_with_llm_ignores_out_of_range_relative_ids(monkeypatch, c
[extracted],
indexes,
state,
ensure_ascii=False,
episode=_make_episode(),
previous_episodes=[],
entity_types=None,
@ -428,7 +425,6 @@ async def test_resolve_with_llm_ignores_duplicate_relative_ids(monkeypatch):
[extracted],
indexes,
state,
ensure_ascii=False,
episode=_make_episode(),
previous_episodes=[],
entity_types=None,
@ -470,7 +466,6 @@ async def test_resolve_with_llm_invalid_duplicate_idx_defaults_to_extracted(monk
[extracted],
indexes,
state,
ensure_ascii=False,
episode=_make_episode(),
previous_episodes=[],
entity_types=None,
@ -498,7 +493,6 @@ async def test_extract_attributes_without_callback_generates_summary():
episode=episode,
previous_episodes=[],
entity_type=None,
ensure_ascii=False,
should_summarize_node=None, # No callback provided
)
@ -529,7 +523,6 @@ async def test_extract_attributes_with_callback_skip_summary():
episode=episode,
previous_episodes=[],
entity_type=None,
ensure_ascii=False,
should_summarize_node=skip_summary_filter,
)
@ -560,7 +553,6 @@ async def test_extract_attributes_with_callback_generate_summary():
episode=episode,
previous_episodes=[],
entity_type=None,
ensure_ascii=False,
should_summarize_node=generate_summary_filter,
)
@ -595,7 +587,6 @@ async def test_extract_attributes_with_selective_callback():
episode=episode,
previous_episodes=[],
entity_type=None,
ensure_ascii=False,
should_summarize_node=selective_filter,
)
@ -605,7 +596,6 @@ async def test_extract_attributes_with_selective_callback():
episode=episode,
previous_episodes=[],
entity_type=None,
ensure_ascii=False,
should_summarize_node=selective_filter,
)